To explore a unique approach for generating robust long term immune responses at both mucosal and systemic lymphoid tissues in the nonhuman primate model for human immunodeficiency virus (HIV), we propose to utilize an attenuated rhesus cytomegalovirus (RhCMV) vaccine vector developed in our laboratories for oral delivery of a simian immunodeficiency virus (SIV) immunogen engineered as a fusion protein to include either full length coding sequence or specific domains of flagellin, the agonist for toll-like receptor (TLR)5. TLR5 is a potent activator of mucosal immune responses through activation of dendritic cells (DC), follicle-associated epithelium (FAE) and M cells included within gut and oral cavity mucosal- associated lymphoid tissue (MALT). As such, flagellin has proven to be a potent adjuvant for i) induction of serum and/or secretory antibody responses and for ii) induction of Th1 and Th2 responses in both systemic and mucosal tissues. This oral vaccine approach combines the advantages of a potent TLR adjuvant with a viral vector (CMV) shown to disseminate throughout the host after oral delivery/transmission and to establish a lifelong persistence in the presence of immune responses that limit viral pathogenicity at multiple sites within the host. Furthermore, our attenuated RhCMV vector has been modified by deletion of the RhCMV-encoded viral interleukin (IL)-10 gene (RhCMV IL10) that results in increased inflammation at the site of primary infection in vivo, an abrogation of the inhibition of DC maturation associated with wild type CMV infection in vitro, and attenuated viral replication for safety. Lastly, data has shown that prior immunity to human CMV (HCMV) does not preclude re-infection, implying that a CMV-based vector system may be broadly applied to a vaccine target population irrespective of pre-vaccine immunity to HCMV. We hypothesize that RhCMV IL10 vectors expressing SIV vaccine immunogens delivered orally and systemically will induce superior vaccine antigen-specific cellular and humoral immune responses in multiple mucosal sites. Also, incorporation of TLR5 agonist adjuvant activity by fusion of flagellin sequences with vaccine antigen will further enhance both innate and adaptive mucosal immune responses and provide a novel oral mucosal vaccine approach for protection against HIV. Accordingly, immunogens (SIVmac239 Capsid) fused with full length and/or specific domains of flagellin will be constructed to generate a vaccine antigen with optimal immunogenicity for mucosal tissues and will be used for generation of candidate RhCMV IL10Capsid/flagellin vaccine vectors (Aim1). Candidate RhCMV IL10Capid/flagellin vaccines will next be tested for immunogenicity (Aim 2) in multiple mucosal and systemic tissues and for efficacy (Aim 3) in RhCMV-seronegative female rhesus macaques by vaginal SIVmac251 challenge. We propose that these novel vectors carry the potential to significantly impact HIV vaccine design.